Data Analysis of ESG Stocks in the Chinese Stock Market Based on Machine Learning

Abstract

In recent years, “environmental, social, and governance” (ESG) stocks have become the best investment considerations. This article selects data from the ESG 300 Index (931463.CSI) and ESG 300 Index (399378.SZ) of the Chinese stock market and analyze its volatility and stock performance from April 2020 to September 2021. We compare the Sharpe Ratio, Sotino Ratio, Treynor Ratio, and Information Ratio of the ESG 300 Index with the Shanghai Stock Exchange and Shenzhen Stock Exchange Composite Index. We use GARCH (1,1) model to assess the volatility and apply machine learning technical skills, including KNN, SVM, and AdaBoost algorithm, to identify the relationship between ESG scores and stock returns collected from RANKING CSR Rating Agency. Based on different ESG scores, we construct hedge fund portfolios and separate them into four classes: Excellent ESG portfolios, Good ESG portfolios, Fair ESG portfolios, and Bad ESG portfolios, and compare with the Sharpe ratio under the situation of the effective tangent frontier. The results show that, compared with non-ESG-related stocks, ESG-related stocks cannot bring excess returns and have better risk performance during a normal period.